In video compression methods such as MPEG-1, MPEG-2, MPEG-4 H.264/MPEG-4 Advanced Video Coding (AVC), an image is split into blocks of a predetermined size and then residual data of each block is obtained by using inter prediction or intra prediction. The residual data is compressed through transformation, quantization, scanning, run length coding, and entropy coding. In entropy coding, syntax elements, e.g., information such as a transformation coefficient and a prediction mode, are entropy-encoded and output in a bitstream. A decoder extracts the syntax elements by parsing the bitstream, and reconstructs the image based on the extracted syntax elements.